Yasantha Niroshan

Graduate Research Assistant · University of Moratuwa

I work on machine learning under tight resource budgets.

I hold a B.Sc. (Hons) in Computer Science and Engineering from the University of Moratuwa, where I'm now a Graduate Research Assistant working on predictive monitoring. I was previously Lead Software Engineer at RoboticGen.

News

Sep 2026I served as reviewer for the GlobalSouthAI workshop at NURIPs.
Sep 2026Our final year project "Predicting Atrial Fibrillation at Edge" has been selected for the Final Round of the SLAAI Best AI Undergraduate Project Award 2026.
Aug 2026Paper "Patient Aware Contrastive Learning for Paroxysmal Atrial Fibrillation Detection Using RR Intervals" nominated for Best Paper at MERCon (IEEE).

Show 5 more

Aug 2026Taking a career break from my role as Lead Software Engineer at RoboticGen to focus on building my research career.
Jul 2026Started as a Graduate Research Assistant at the University of Moratuwa, working on predictive maintenance.
Jul 2026Served as a reviewer for the GlobalSouthAI Workshop at IJCAI 2026.
May 2026New workshop paper, "Patient-Aware Contrastive Learning...", accepted at the ICML GlobalSouthML Workshop.
Apr 2026Released the Bee Acoustic Dataset with Environmental Parameters on Zenodo.

Research Interests

Embedded Machine Learning (TinyML)
Neuromorphic Computing
Resource-Constrained Neural Network Optimization
Efficient Edge Inference

Selected Publications

Patient-Aware Contrastive Learning Preserves Per-Patient Structure in RR-Interval Representations

Yasantha Niroshana, Weijith Wimalasiri, Chathuranga Hettiarachchi

GlobalSouthML Workshop, International Conference on Machine Learning (ICML) · 2026

arXiv Code

BibTeX

@inproceedings{niroshana2026patientaware,
  title     = {Patient-Aware Contrastive Learning Preserves Per-Patient Structure in RR-Interval Representations},
  author    = {Niroshana, Yasantha and Wimalasiri, Weijith and Hettiarachchi, Chathuranga},
  booktitle = {GlobalSouthML Workshop, International Conference on Machine Learning (ICML)},
  year      = {2026},
  note      = {arXiv:2606.23570}
}

Patient Aware Contrastive Learning for Paroxysmal Atrial Fibrillation Detection Using RR Intervals

Yasantha Niroshana, Weijith Wimalasiri, Joshua P. Kulasingham, Chathuranga Hettiarachchi

Moratuwa Engineering Research Conference (MERCon), IEEE — Best Paper Award Nominee · 2026

Peak detection of PPG signals using fixed-point digital filters implemented in VHDL

H. K. Y. Niroshana, W. M. Wimalasiri, C. Hettiarachchi

ERU Symposium, University of Moratuwa · 2025

Paper DOI Code

BibTeX

@inproceedings{niroshana2025ppg,
  title     = {Peak detection of {PPG} signals using fixed-point digital filters implemented in {VHDL}},
  author    = {Niroshana, H. K. Y. and Wimalasiri, W. M. and Hettiarachchi, C.},
  booktitle = {Proceedings of the Engineering Research Unit (ERU) Conference},
  year      = {2025},
  publisher = {Engineering Research Unit, University of Moratuwa},
  doi       = {10.31705/ERU.2025.38},
  url       = {https://dl.lib.uom.lk/items/73c999d1-3082-4812-94ae-30e70bf6d68a}
}

All publications → · Full CV →